Habitat loss and climate change together represent a great threat to biodiversity because species face the difficult task of shifting their distributions across human-dominated landscapes in which suitable habitats are present only as scattered fragments of formerly more widespread types of vegetation. New approaches are needed to understand and predict accurately the responses of species to these environmental drivers of change when they act in combination. The complication is that climate change itself alters habitat quality and quantity, by changing the availability of suitable 'microclimates'. For example, our study species, the silver-spotted skipper butterfly, reaches the cool northern edge of its European distribution in England. As such, it used to be restricted to exceptionally hot microclimates (short grass on South-facing hillsides) in the early 1980s, but has recently colonised cooler habitats (taller grassland, and East, West, and North-facing hillsides) as the climate has warmed. Habitats that used to be too cool are now accessible to the species; although some south-facing grassland may have started to become too hot and dry for the insect. Increasing suitability of East, West, and North-facing hillsides has resulted in a major increase in the amount of grassland that is thermally acceptable, allowing the silver-spotted skipper to start to expand its distribution. However, the situation is complicated because year-to-year variation in climatic conditions constantly alters the suitability of each remaining area of calcareous grassland (depending on the slope, aspect and vegetation). The habitat available to the skipper is a shifting mosaic depending on the weather conditions each year, making it difficult to provide clear guidelines for conservation managers to allow the species to survive and extend its distribution. As the climate changes, this interaction between climate and habitat is likely to complicate the process of conservation planning and habitat management for the many rare species that are now restricted to localised areas of habitat in modern landscapes. To date, the feedback loop between climate and the landscape-scale distribution of habitat has not been incorporated in any scientific modelling framework, but this is required before believable and testable projections of species responses to climate change can be made. We will develop a new approach using a population model that incorporates variation over time in climate-driven habitat availability. These models will be developed using large-scale data on the British distribution, habitat and population sizes of the silver-spotted skipper butterfly for the period 1982 to 2001. The models will then be used to predict post-2001 changes, and we will test our projections against new information on changes in habitat and distribution for the skipper between 2002 and 2010. The project will allow us to test how accurately we can predict changes in species distributions as they respond to climate change, and the importance of climate, habitat and their interactions in explaining the rates at which species extend their distributions. This step is vital to determine whether conservation actions can alleviate the effects of climate change on biodiversity, and which actions are most efficient in this process of adapting conservation to climate change. We will make the software that we develop available to other scientists, policy-makers and conservation practitioners, allowing our approach to be applied to the conservation of the many other rare species facing the same problems as the silver-spotted skipper.